Model: GELU Model: 4 Layers, 2048 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 487 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.3050. Min Range: -3.3050

Max Act: 3.3050. Min Act: -0.1700

Data Index: 576451 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 3.3050. Min Range: -3.3050

Max Act: 3.0847. Min Act: -0.1700

Data Index: 1225213 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 3.3050. Min Range: -3.3050

Max Act: 3.0689. Min Act: -0.1700

Data Index: 1302972 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 3.3050. Min Range: -3.3050

Max Act: 3.0847. Min Act: -0.1699

Data Index: 384730 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 3.3050. Min Range: -3.3050

Max Act: 3.0373. Min Act: -0.1700

Data Index: 1550993 (Python Code)

Max Activating Token Index: 1023

Text #5

Max Range: 3.3050. Min Range: -3.3050

Max Act: 3.0847. Min Act: -0.1700

Data Index: 170704 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 3.3050. Min Range: -3.3050

Max Act: 3.0215. Min Act: -0.1700

Data Index: 1654833 (Python Code)

Max Activating Token Index: 1023

Text #7

Max Range: 3.3050. Min Range: -3.3050

Max Act: 3.0057. Min Act: -0.1700

Data Index: 193525 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 3.3050. Min Range: -3.3050

Max Act: 2.9741. Min Act: -0.1700

Data Index: 1447686 (Python Code)

Max Activating Token Index: 1023

Text #9

Max Range: 3.3050. Min Range: -3.3050

Max Act: 3.0057. Min Act: -0.1700

Data Index: 1386719 (Python Code)

Max Activating Token Index: 1023

Text #10

Max Range: 3.3050. Min Range: -3.3050

Max Act: 2.9741. Min Act: -0.1700

Data Index: 1467241 (Python Code)

Max Activating Token Index: 1023

Text #11

Max Range: 3.3050. Min Range: -3.3050

Max Act: 2.9899. Min Act: -0.1700

Data Index: 1435816 (Python Code)

Max Activating Token Index: 1023

Text #12

Max Range: 3.3050. Min Range: -3.3050

Max Act: 3.0215. Min Act: -0.1700

Data Index: 1268393 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 3.3050. Min Range: -3.3050

Max Act: 3.0057. Min Act: -0.1700

Data Index: 735527 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 3.3050. Min Range: -3.3050

Max Act: 2.9741. Min Act: -0.1700

Data Index: 925397 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 3.3050. Min Range: -3.3050

Max Act: 3.0057. Min Act: -0.1700

Data Index: 1118735 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 3.3050. Min Range: -3.3050

Max Act: 2.9899. Min Act: -0.1700

Data Index: 104149 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 3.3050. Min Range: -3.3050

Max Act: 2.9741. Min Act: -0.1700

Data Index: 122056 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 3.3050. Min Range: -3.3050

Max Act: 2.9583. Min Act: -0.1700

Data Index: 1554273 (Python Code)

Max Activating Token Index: 1023

Text #19

Max Range: 3.3050. Min Range: -3.3050

Max Act: 2.9424. Min Act: -0.1700

Data Index: 1168481 (C4 (Web Text))

Max Activating Token Index: 1023